| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Gastrointestinal Imaging |
1 From the Departments of Radiology (P.R., A.L., E.H., M.B., N.E.R., B.H.) and General, Vascular, Thoracic, and Visceral Surgery (J.C.R.), Charité Campus Mitte, Universitätsmedizin Berlin, Schumannstr 20/21, 10098 Berlin, Germany. Received January 7, 2004; revision requested March 9; final revision received August 8; accepted August 25. Address correspondence to P.R. (e-mail: rogalla{at}charite.de).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MATERIALS AND METHODS: This study had institutional review board approval; subject informed consent was not required. CT colonography was performed without the administration of an intravenous spasmolytic in 80 asymptomatic subjects (group 1; 45 women, 35 men; age range, 48–77 years; mean, 61.9 years). These subjects were matched with two groups of 80 subjects who were similar in age but were premedicated with glucagon (group 2; 41 women, 39 men; age range, 43–76 years; mean, 63.1 years) or butyl scopolamine (group 3; 43 women, 37 men; age range, 34–77 years; mean, 63.4 years). All 240 subjects were examined in the supine position with multisection CT and a section thickness of 1 mm after intravenous contrast agent administration and rectal carbon dioxide insufflation. The colon was divided into seven segments, and the colon length, total volume, radial distensibility, and number of nondistended segments were calculated for each subject and compared among the three groups. Statistical analysis was performed with analysis of variance and 
2 testing.
RESULTS: Mean bowel length was not significantly different among the groups. Mean colon volumes and radial distensibilities, respectively, were 1.84 L and 3.69 cm in group 1, 2.14 L and 3.98 cm in group 2, and 2.35 L and 4.23 cm in group 3; differences in colon volume and radial distensibility were significant only between group 1 and group 3 (P < .001). At CT colonography, 29 segments in 20 group 1 subjects were collapsed, 23 segments in 12 group 2 subjects were collapsed, and 11 segments in six group 3 subjects were collapsed (P = .016).
CONCLUSION: Premedication with butyl scopolamine or, less effectively, glucagon improves colonic distention in the supine subject.
© RSNA, 2005
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
For comparison, two groups of 80 asymptomatic subjects who had been examined during the same time period were selected from among a total of 1239 subjects who had undergone CT scanning with an identical imaging protocol. The two groups differed solely in the type of spasmolytic drug administered: Either 1 mg of glucagon (Glucagen; Novo Nordisk Pharma, Mainz, Germany) had been used (study group 2; 41 women [age range: 49–76 years; mean age, 63.4 years] and 39 men [age range, 43–72 years; mean age, 62.7 years] with a mean age of 63.1 years) or 20 mg of butyl scopolamine (Buscopan; Boehringer Ingelheim Pharma, Ingelheim am Rhein, Germany) had been used (study group 3; 43 women [age range, 34–76 years; mean age, 64.3 years] and 37 men [age range, 48–77 years; mean age, 62.5 years] with a mean age of 63.4 years). Each drug had been slowly injected intravenously 2 minutes before the initiation of CT scanning.
At our institution, subjects undergoing CT colonography routinely receive butyl scopolamine; however, if contraindications are present, glucagon is used. The criterion used to identify each patient in both matched groups was the age ± 3 years of a group 1 subject. In an attempt to avoid bias stemming from the manual selection procedure, we sequentially selected subjects by counting backward starting from the examination date of the last subject included in study group 1. For bowel cleansing, subjects had to drink a phosphosoda preparation (a 45-mL solution of bisodium hydrogen phosphate) together with 30 mL of an iodinated oral contrast agent (Peritrast; Koehler Chemie, Alsbach-Haenlein, Germany) after lunch on the day before the examination.
A second dose of the phosphosoda preparation, together with 30 mL of the iodinated oral contrast agent, was taken in the evening instead of dinner. After skipping breakfast, the subjects arrived for the CT examination the next morning. No further dietary modifications were required. After informed consent for the clinical examination was obtained and a 22-gauge venous line (Braunüle Vasocan; B. Braun Melsungen, Melsungen, Germany) was placed, each subject was positioned on the CT table. A rectal tube (E-Z-Em, Westbury, NY) was then inserted and connected to a carbon dioxide gas bottle. The tube was equipped with a clamp and a lateral outlet for manual release of the gas flow during insufflation. With the subject lying in the supine position, a trained nurse gently insufflated the colon manually until the subject experienced discomfort.
A CT scout scan was acquired for planning the CT study and assessing the degree of bowel distention. During the scan delay after initiation of intravenous contrast medium administration, the colon was further insufflated as far as the subject could tolerate. Multisection helical CT scanning was performed during maximal inspiration. Complete distention of all colonic segments was checked by a radiologist during the examination on the basis of real-time reconstruction of the CT data. When a segment was not distended (ie, when it was collapsed), the subject was turned to the prone position immediately after the breathing instruction (expiration command) was given, and the affected segment was scanned again by using the same CT parameters, without renewed contrast medium injection or gas insufflation. After the CT examination was completed, subjects were routinely kept in the holding area for 10 minutes so that they could be observed for possible side effects caused by the intravenous medications.
Imaging
Examinations were performed by using a four–detector row multisection CT scanner (Aquilion; Toshiba Medical Systems, Tochigi, Japan) and the following parameters: 1-mm collimation, an 0.8-mm reconstruction interval, a 6.5:4 pitch factor, 120 kV, 110 mAs, and a 0.5-second gantry rotation time. The CT examination was begun with a delay of 60 seconds after intravenous administration of 120 mL of iopromide (Ultravist 370; Schering, Berlin, Germany) at a flow rate of 2.5 mL/sec.
Image Evaluation
The CT data sets of all examinations were postprocessed to create endoscopic images on a UNIX-based workstation (EasyVision Release 5.1.2; Philips Medical Systems, Eindhoven, the Netherlands). So that we could determine the colonic volume and compute a path through the colon (with the subject in the supine position), the starting point of the path was manually defined (by P.R. [who had 7 years of experience with CT colonography] and either E.H. [who had 3 years of experience] or A.L. [who had 2 years of experience]) as the center of the cecum, and the end point was selected in the distal portion of the rectum. The computer then calculated the path through the colon; the corresponding colon length could be directly read in centimeters. In the presence of stenosis or when a bowel segment was collapsed, the path had to be traced manually by the reading radiologists.
After the path was computed, the colon was reconstructed by using a classification of –400 HU and a gradient of 20 HU, both as a nontransparent and as a transparent model; this enabled automatic determination of the volume in millimeters. Given simplification of the colon as a cylindric model, the average radial distensibility, or RD, of the colon was calculated in centimeters according to the following formula:
 |
|
2 test was used for comparison of differences in subject sex. The number of collapsed colon segments was compared both on a per-subject and on a per-segment basis (80 subjects times seven segments equals 560 segments per group), and differences between the groups were tested for significance by using the 2 test, with which significance was accepted when P was less than .017 (the Bonferroni correction for stacked comparisons with original significance levels of P < .05 was applied). All tests were performed with GraphPad InStat, version 3.05 (GraphPad Software, San Diego, Calif) for Windows. |
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
|
The total difference among all three groups at per-subject analysis reached a significance level of P = .0024, with P = .17 (not significant) for group 1 versus group 2, P = .0053 (significant) for group 1 versus group 3, and P = .21 (not significant) for group 2 versus group 3. A per-segment analysis revealed similar degrees of significance, with P = .016 for all three groups, P = .037 (not significant) for group 1 versus group 2, P = .004 (significant) for group 1 versus group 3, and P = .39 (not significant) for group 2 versus group 3.
For 46 (73%) of the 63 segments that were collapsed among all three groups, at least one of the two sigmoid segments were involved; for 10 (16%) of the 63 segments, the descending colon was involved; for four (6%) of the 63 segments, the transverse colon was involved; and for three (5%) of the 63 segments, the rectum was involved. All subjects with initially collapsed bowel segments had adequate distention of the segments when CT scanning was repeated with subjects in the prone position; this fact enabled complete evaluation of the colonic lumen on the basis of a combined reading of both data sets. In no subject was colorectal cancer or severe acute or chronic diverticulitis diagnosed. Thirteen subjects in group 1, nine subjects in group 2, and 16 subjects in group 3 had diverticular disease without signs of inflammation. The lumen, however, was sufficiently distended throughout the affected parts of the colon in these subjects.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Both butyl scopolamine and glucagon have been used as effective spasmolytic agents for conventional diagnostic assessments of the bowel for many years (10). Compared with intramuscular injection (an alternative route of administration for conventional radiography) of these agents, intravenous administration of these agents is followed within a few seconds by the onset of action, but the effect lasts for only a few minutes. The use of butyl scopolamine is limited by its contraindications, which include prostatic hyperplasia and clinically relevant tachycardia; increased intraocular pressure is a relative contraindication (11). Patients must be questioned about the presence of these contraindications and informed about the risks of use of the drug. However, the most important disadvantage of butyl scopolamine appears to be that it reduces the adaptability of the eyes and that subjects should not drive an automobile after its administration. This may be a problem for outpatients. Aside from its effects on glucose metabolism, the alternative preparation, glucagon, has fewer contraindications (hypersensitivity, pheochromocytoma), but it is more expensive than butyl scopolamine. The administration of either substance appears to be justified only when there is a diagnostic benefit for CT colonoscopy.
Many physicians advocate that CT colonography be performed with patients in both the supine and the prone position (12–17). The aim of dual positioning is to achieve complete distention of all bowel segments by redistributing intraluminal air and to enable visualization of the entire intestinal wall in contrast to air by redistributing any residual fluid or stool. Residual stool adhering to the wall may mimic a polyp, but such stool should appear in a different place in the second study performed after the patient is repositioned and can thereby be distinguished from a true polyp. Single positioning may prevent readers from gaining knowledge about the redistribution of stool or fluid; however, the major disadvantages of nonselective repeat scanning are that it results in additional discomfort and a double radiation dose to the patient, twice the number of images to be interpreted, the need for increased disk storage capacity, and a longer examination time. For these reasons, it seems more appropriate to restrict repeat scanning to those bowel segments that are collapsed in the prone position. However, such a procedure requires rapid or even real-time reconstruction of the CT images and that a radiologist be present during the scanning procedure. Another approach was employed by Pineau et al (18), who reexamined only the pelvic region with patients in the prone position after a complete abdominal study was performed with patients in the supine position.
Excessive residual endoluminal fluid is avoided by oral administration of a contrast medium (Peritrast in the present study) together with a laxative (19). The oral contrast medium has a moderate cathartic effect that acts synergistically with phosphosoda and is tolerated well. As a result of the contrast effect, residual fluid in the bowel appears bright, and the bowel wall is delineated from this fluid by negative contrast (18). Residual stool likewise takes up the contrast medium and can thus be differentiated from polyps (fecal tagging); additional ingestion of barium sulfate as part of a clear-liquid diet has also been successfully used in a large study group (3).
In the present study, the volume and radial distensibility of the colon were significantly higher after premedication with butyl scopolamine than without premedication; likewise, the proportion of collapsed bowel segments was lower in the butyl scopolamine group than in the other two groups. These observations confirm the subjectively better image quality noted in subjects who received butyl scopolamine. This effect is most likely attributable to a delayed onset of distention pain during insufflation due to smooth muscle relaxation, which results in a greater insufflation volume.
It was not the aim of the present study to prove that butyl scopolamine increases the sensitivity of CT colonography in colorectal polyp detection. However, our results show that butyl scopolamine and, less effectively, glucagon improve rectal filling of the colon with carbon dioxide. With bowel distention being a prerequisite for assessment of the bowel wall, our findings suggest that the use of butyl scopolamine (and, to a lesser extent, glucagon) has the potential to improve the diagnostic quality of CT colonoscopy if only supine positioning of the patients is used.
Butyl scopolamine has not yet been approved for use in patients in the United States. Morrin et al (20) investigated the effect of glucagon in 74 patients who were compared with 22 patients who had not received premedication, and they did not find any beneficial effect of this agent in CT colonography. Subjective assessment revealed distention of 85.2% of bowel segments in the group with premedication, compared with 84% of segments in the group without premedication. Similar results were reported by Yee et al (21), who observed no improvement in colonic distention after glucagon administration in a study of 60 patients, 33 of whom had received 1 mg of glucagon intravenously and 27 of whom had received no spasmolytic agent. However, all patients had received magnesium citrate and polyethylene glycol, both of which leave more fluid in the bowel than phosphosoda (22,23).
The results of our study demonstrate that the use of glucagon had a significant effect with respect to a reduction in the number of collapsed segments when it was tested alone against the use of no premedication; however, this effect was less pronounced when the effect of using glucagon was compared with the effect of using butyl scopolamine. Regarding colonic volume after insufflation, our data support the results of the above-cited studies; furthermore, because all segments in all subjects could ultimately be evaluated at a combined reading of both the supine and selected prone scans, dual scanning appears to be effective, as previously concluded by Bruzzi et al (24). However, our data do not support the conclusion of Morrin et al (20) that spasmolytics other than glucagon likewise have only a small effect in improving bowel distention.
At approximately 
26 (about $33) versus approximately 5 (about $6) per patient, glucagon is much more expensive than butyl scopolamine (10). Because butyl scopolamine is more effective than glucagon in reducing segmental collapse, one should always carefully consider the individual patient's clinical findings and history to determine whether intravenous spasmolysis with butyl scopolamine is actually contraindicated. Reports in the literature suggest that increased intraocular pressure, in particular, is overrated as a contraindication (11). The need to clearly inform all patients about the risk of temporary visual disturbances that may occur after administration of butyl scopolamine should not be forgotten. Another disadvantage of the use of any intravenous premedication is the fact that venous access is required, even when an unenhanced CT study is performed.
Administration of intravenous contrast material (25), although not typically recommended (26), and the use of a double-dose phosphosoda cleansing regimen are part of the routine protocol for CT colonography at our institution. The use of a double-dose phosphosoda cleansing regimen carries a U.S. Food and Drug Administration warning and is generally not recommended in the U.S.; however, a double-dose preparation regimen is commonly used in most European countries and in Asia and has been evaluated for its effectiveness (27). Because the same bowel preparation technique was used for all subjects in the three groups in the present study, it appears unlikely that the type of cleansing influenced the bowel distention results.
One limitation of our study lies in the potential bias related to the selection of the subjects: Because the subjects in group 1 had contraindications to both glucagon and butyl scopolamine or refused premedication, it is possible that these subjects tolerated colonic distention less well than did subjects who had no contraindications and were willing to receive premedication. We also did not assess the amount of gas that refluxed into the small bowel. One might assume, however, that after administration of muscle relaxants, a larger amount of gas could pass the ileocecal valve. Given that the premedicated groups had a larger colon volume, reflux of gas into the small bowel is not likely to have influenced our results.
Although we did not assess subject comfort during the examination, it is possible that, because spasmolytics result in a delayed onset of pain or an overall reduction in distention pain, administration of muscle relaxants potentially improves patient acceptance of the procedure. Our overall data suggest that the use of intravenous spasmolysis with butyl scopolamine can be recommended for CT colonography when supine patient positioning is preferred.
|
FOOTNOTES |
|---|
Author contributions: Guarantors of integrity of entire study, P.R., A.L.; study concepts, P.R., A.L., M.B.; study design, P.R., M.B., J.C.R.; literature research, N.E.R.; clinical studies, P.R., A.L., E.H.; data acquisition, P.R., A.L.; data analysis/interpretation, P.R., A.L., J.C.R.; statistical analysis, N.E.R., A.L.; manuscript preparation, N.E.R., E.H.; manuscript definition of intellectual content, P.R., B.H.; manuscript editing, N.E.R.; manuscript revision/review, P.R., N.E.R., E.H.; manuscript final version approval, P.R., N.E.R., B.H.
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  D. S. WERDERMAN CT Colonography Radiol. Technol., March 1, 2007; 78(4): 309CT - 324CT. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Mang, A. Maier, C. Plank, C. Mueller-Mang, C. Herold, and W. Schima Pitfalls in Multi-Detector Row CT Colonography: A Systematic Approach RadioGraphics, March 1, 2007; 27(2): 431 - 454. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
